基于灰色RBF组合模型的宁波港集装箱海铁联运量预测

Prediction of Container Sea-rail Intermodal Transport Volume at Ningbo Port Based on Grey-RBF Neural Network Combination Model

集装箱海铁联运量预测是海铁联运网络规划设计的重要组成部分,影响港口及其集疏运体系进一步发展,因此,对集装箱海铁联运量做出科学合理的预测显得十分重要.以2009~2015年宁波港集装箱海铁联运量为原始数据,运用灰色RBF神经网络组合模型预测其未来2年集装箱海铁联运量增长趋势.预测结果表明,灰色RBF神经网络组合模型预测精度高于GM（1,1）模型、RBF神经网络模型、灰色BP神经网络组合模型.可见,该组合模型可有效应用于集装箱海铁联运量的预测领域.

It is of great importance to make reasonable predictions on container sea-rail intermodal transport due to the fact that it will affect the future development of port and transportation system as an important element in planning and designing. Based on the real-time data of container in Ningbo port collected from 2009 to 2014, the paper predicts the growth trend of container sea-rail interrnodal transport volume in the coming two years by employing Grey-RBF neural network model. Prediction results show that the Grey-RBF neural network model is superior in accuracy than GM （1,1） model, the RBF neural network model and the Grey-BP neural network combined model. The effectiveness of predictions is also validated in this paper.